| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J. Biol. Chem., Vol. 281, Issue 30, 21131-21138, July 28, 2006
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
From the Department of Physiology, University of Texas Health Science Center, San Antonio, Texas 78229
Received for publication, March 31, 2006 , and in revised form, May 23, 2006.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
It was hypothesized that the force that opens the TM2 "gate" comes from a Ca2+-dependent conformational change in the RCK domains, which in turn tugs on a linker segment that is directly connected to TM2 (1). To be consistent with the principle underlying allosteric modulation of the channel, one would predict that in the Ca2+-bound conformation, the "splaying apart" of the TM2 segments (which opens the channel) is clearly energetically favored. TM2 can also presumably bend to close the channel while Ca2+ is bound, just as the channel can open while the channel is not Ca2+-bound, although the closed state is favored. In principle, the intrinsic equilibrium between the two primary conformations of TM2 would not depend directly on Ca2+ and could be modulated by other interactions, such as interactions among side chains near the TM2 bundle crossing, as observed with mutants of the Shaker K+ channel (6, 7).
To look for potential interactions that may modulate the intrinsic MthK gating equilibrium, we used a bacterial complementation strategy. Using Escherichia coli strains that are deficient in K+ uptake, we identified a series of mutations that decreased complementation (and thus K+ uptake) in the strains. We then characterized the mutants using biochemical and electrophysiological assays. Our studies demonstrate that wild-type MthK is complementary in K+ uptake-deficient E. coli and thus supports K+ uptake via its open pore, whereas several mutations near the putative TM2 bundle crossing can reduce or eliminate complementation, primarily by reducing channel open probability. Because spontaneous channel opening is reduced in these mutants at very low Ca2+, as indicated by our complementation assay, and Ca2+-dependent opening is reduced at higher Ca2+, as shown in our electrophysiological recordings, our studies suggest that the negative charge at Glu-92 near the putative MthK bundle crossing may be critical in determining the relative stability of the open conformation of MthK.
|
|
EXPERIMENTAL PROCEDURES |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
trkA kup1 (trkD1) kdpABC5 rpsL metE thi rha gal) (8) was kindly provided by Evert Bakker (University of Osnabruck, Germany). MthK wild-type cDNA (1), kindly provided by Christopher Miller (HHMI, Brandeis University), was obtained in the pQE-70 vector, subcloned into the pQE-82L vector (Qiagen) between the SphI and BglII restriction sites, and then modified by inserting a thrombin-cleavable His6 tag at the C-terminal end of the gene and deleting the N-terminal His6 tag from the pQE-82L vector. Site-specific mutants were generated using QuikChange (Stratagene) and confirmed by sequencing. Complementation Assay—Complementation was assayed as described previously (9), except that the LB2003 strain was used in the present studies. Briefly, bacteria were transformed with plasmid DNA and grown overnight in high K+ medium (KLM) containing 5 g/liter yeast extract, 10 g/liter tryptone, 10 g/liter KCl, and 100 mg/liter ampicillin. From the overnight culture, 100 µl was diluted in 5 ml of KLM and grown to A600 = 0.5. Channel expression was induced by the addition of 1 mM IPTG. After 2 h of induction, the cell density was measured, and the culture was diluted (normalized) to an A600 of 0.5. A 1-ml sample of this suspension was pelleted and stored at -20 °C to be used in SDS-PAGE and Western blot analysis. 3.5-µl drops of the diluted cell suspension were spotted on ampicillin plates containing 5 g/liter yeast extract, 10 g/liter tryptone, and either 100 mM KCl with 15 mM NaCl or 1 mM KCl with 114 mM NaCl, with or without IPTG. Reproducibility of complementation results was confirmed in at least three different experiments for each mutant.
To assay bacterial growth in liquid medium, the culture density of induced transformants was normalized as described above, centrifuged (10,000 x g for 2 min), washed once, and used as innoculum for growth in LB with 2 mM KCl, ampicillin, and IPTG.
Western Blot—Pelleted samples of induced transformants (see above) were resuspended in a lysis buffer containing 20 mM Tris-HCl, 4 M urea, 2% (w/v) SDS, pH 6.0. Lysates were centrifuged at 10,000 x g for 20 min, and 10 µl of the supernatant was mixed with 10 µl of 2x Laemmli sample buffer and loaded into a 10% polyacrylamide gel. After separation by SDS-PAGE, protein was transferred to a nitrocellulose membrane using a wet transfer apparatus (Bio-Rad). The membrane was washed with Tris-buffered saline (TBS) and incubated in blocking buffer (TBS with 0.1% Tween 20 and 5% (w/v) nonfat dry milk) for 60 min. Penta-His monoclonal antibody (Qiagen) was added followed by incubation overnight at 4 °C. After a 30-min incubation at room temperature, the membrane was washed three times with TBS and incubated with peroxidase-labeled goat anti-mouse IgG (Kirkegaard and Perry) in blocking buffer for 60 min at room temperature. The membrane was then washed three times, and antibody was visualized using peroxidase-3,3'-diaminobenzidine substrate (DAB substrate kit, Vector Laboratories).
Channel Purification and Reconstitution—MthK was expressed and purified essentially as described previously (1). Briefly, MthK wild-type and mutants were expressed in E. coli XL-1 Blue cells on induction with 0.4 mM IPTG. Bacteria were harvested, resuspended in 20 mM Tris, 100 mM KCl, pH 7.6 (Buffer A), and lysed by sonication in the presence of phenylmethylsulfonyl fluoride and a protease inhibitor mixture (Complete EDTA-free, Roche Applied Science). The protein was solubilized by 2 h of incubation in Buffer A with 50 mM decyl maltoside (DM) (Anatrace) followed by centrifugation at 16,000 rpm for 45 min. The supernatant was loaded onto a Talon Co2+ affinity column (Clontech) and washed with 20 mM imidazole and 5 mM DM in Buffer A, and the channel protein was eluted using 400 mM imidazole and 5 mM DM in Buffer A. The His6 tag was cleaved immediately after elution by incubating with 2.0 units of thrombin/3.0 mg of eluted protein for 2 h at room temperature. Protein was concentrated to a volume of 1 ml using an Amicon Ultra filter (molecular weight cutoff 10,000), further purified on a Superdex-200 gel filtration column, and then concentrated again and reconstituted into liposomes composed of E. coli lipids (Avanti). Protein concentrations in liposomes ranged from 5 to 50 µg of protein/mg of lipid.
|
10 µM, as measured with a Ca2+-sensitive electrode (Orion Research). Orientation of mutant channels in the bilayer membrane was typically identified using the intrinsic inward rectification properties of the channels (1). Electrophysiological data (open probability and unitary conductance measurements) are expressed as mean ± S.E. from at least three different experiments for each data point.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
We observed that the MthK channel complements bacterial survival in the K+ uptake-deficient LB2003 strain (Fig. 2). This result is consistent with our previous observation that MthK complements survival in the similarly K+ uptake-deficient TK2446 strain (9). In addition, we observed that complementation by MthK requires expression of the pore-forming transmembrane region; expression of the cytoplasmic RCK domain alone does not support K+ uptake. This suggests that the K+ uptake pathway in the transformed bacteria is likely to be the open MthK pore.
MthK Channel Opening Is Increased at Elevated Temperature—The complementation due to MthK wild-type (WT) channel expression (Fig. 2) suggests that the MthK channel opens in vivo, in the E. coli membrane. We were at first surprised by this result, because in lipid bilayer experiments it appears that MthK opening requires millimolar levels of cytoplasmic Ca2+, which is much higher than the levels of cytoplasmic free Ca2+ in E. coli (1, 13). However, previous lipid bilayer recordings were performed at room temperature (1). Thus we hypothesized that at 37 °C, the temperature of the complementation assay, MthK channels may open sufficiently to support K+ uptake and complementation.
|
0.8 in [Ca2+] ranging from 5 to 25 mM (see below and Fig. 7), our results are consistent with two potential mechanisms for the activation of MthK at elevated temperature. Heat could activate the channels either by opening the channel gate directly or by increasing the effective Ca2+-binding affinity by at least 100-fold. Although our experiments did not distinguish between these mechanisms, either of them could lead to MthK opening in E. coli under our culture conditions. We went on to perform mutational screens directed at two regions: the Ca2+-binding site, located in the cytoplasmic RCK domain of the channel, and the TM2 segment, at the cytoplasmic mouth of the MthK pore.
|
To test the effect of knocking out the Ca2+-binding site, we made the isosteric charge-neutral mutations D184N and E210Q. These reduced complementation, as did the double mutation D184N/E210Q (Fig. 4B). However, complementation was not abolished by these mutations, as one might expect if K+ uptake through the channel strictly required Ca2+-dependent activation. Thus the complementation effect may be due in part to spontaneous (Ca2+-independent) opening of the MthK pore.
Mutations at the Putative TM2 Bundle Crossing Can Decrease Complementation—K+ channel opening can be modulated by side chains at the putative gating region of the channel, which is the S6 segment of Kv channels and TM2 helix of KcsA (6, 7, 10). It has been hypothesized that the closed gate of Shaker-type channels can be stabilized by hydrophobic interactions at the S6 bundle crossing (7). Also, replacement of residues near the TM2 bundle crossing of KcsA (A108) with charged side chains results in increased channel opening (10). The residue corresponding to KcsA Ala-108 in MthK is Glu-92, which is negatively charged at neutral pH. We focused on this residue and nearby residues in TM2 to determine their potential roles in stabilizing gate conformation.
Fig. 4 shows the results of complementation analysis of several mutations of Glu-92. To determine the effect of the side chain at the equivalent position in KcsA, we constructed E92A; this resulted in the most striking decrease in complementation. To determine potential side-chain characteristics responsible for the complementation decrease, we made several other substitutions at Glu-92 (Asp, Gln, Val, Pro, and Lys). Although complementation in all cases was reduced in comparison with wild type, only E92P and E92K practically eliminated complementation (like E92A).
In addition to the complementation assay that used survival on agar plates as a qualitative index, we assayed complementation quantitatively in a few mutants using growth in liquid medium. Fig. 4C shows that for the LB2003 cells grown in medium additionally containing 2 mM KCl, cells expressing MthK wild-type were able to grow to an A600 of 1.3 ± 0.02 by 24 h, whereas E92Q, which resulted in reduced complementation on low K+ agar plates, reached an A600 of 0.65 ± 0.05 by 24 h, and E92A reached 0.06 ± 0.01 (reduced by 95%).
Because Glu-92 is adjacent to the positively charged residue Arg-93, we wondered whether altering the Glu-92 side chain would disrupt a key electrostatic interaction between the Glu-92 and Arg-93 side chains required for gating. To test this possibility, we assayed the mutant R93L. R93L yielded complementation that was indistinguishable from wild-type MthK (Fig. 4B), suggesting that a Glu-92—Arg-93 interaction is not critical for gating.
The E92A Mutation Shows a Loss-of-function but Does Not Blunt Expression or Assembly—To test whether effects on complementation were due to altered mutant channel expression, we expressed mutants in LB2003 cells and quantified expression in Western blots of bacterial lysates. Because MthK shows bands corresponding to complete channel assemblies as well as monomeric channel subunits, we were also able to make a crude estimate of the stability of transmembrane channel assembly.
We found that expression levels for most of the mutants mimicked MthK wild-type expression (Fig. 5). Fully assembled channels ran at an apparent molecular mass of 220 kDa (1), although other bands were sometimes observed at 190 and 166 kDa. The 190- and 166-kDa bands appear to represent partially degraded or denatured forms of the channel. We observed that the density of these bands increased with extended storage of the protein in buffers containing 4 M urea, independent of any mutations (data not shown). In addition, a light band was sometimes apparent at 78 kDa, which likely corresponds to a dimeric form of the protein; the presence of this band was also unrelated to any mutations.
In contrast to the high expression levels observed with most of our mutants, E92P showed reduced expression and ran almost entirely in the band corresponding to the monomeric subunit (Fig. 5). If the proline interferes with channel assembly to produce fewer functional E92P channels in E. coli than wild-type MthK, this may account for the decreased complementation by E92P. Neither E92A nor E92K showed a decrease in tetrameric assembly on Western blot compared with the wild type, suggesting that these mutants are expressed and fully assembled in E. coli but cannot support K+ uptake at the same levels as wild-type MthK.
|
|
The E92A/E96T Double Mutation Further Blunts Complementation—The reduced functionality of E92A continued to intrigue us for two reasons. Unlike E92K, E92A does not introduce a positive charge in the pore lining, so it would not necessarily be expected to slow potassium conduction via an electrostatic mechanism. Also, E96A did not reduce complementation, so the alanine substitution seems to be relatively sensitive to the Glu-92 position, unlike the lysine substitution, which reduced complementation at either Glu-92 or Glu-96.
We went on to combine E92A with E96T, which substitutes both of these glutamates in MthK with the side chains present in KcsA at the analogous positions, to determine whether this combination further effected functionality. We found that LB2003 growth in low K+ liquid medium, which was already blunted by the E92A mutation, was nearly abolished in the E92A/E96T double mutant (Fig. 4C). The E92A/E96T double mutation reduced neither the apparent expression levels LB2003 cells nor tetrameric stability in SDS, as indicated by Western blot (Fig. 5) and in Coomassie-stained SDS gels of the purified E92A/E96T protein (Fig. 6). Also, DM-solubilized E92A/E96T eluted with the same elution time as wild-type MthK on gel filtration, further indicating the oligomeric integrity of the mutant (Fig. 6).
|
|
We observed that wild-type MthK required millimolar concentrations of Ca2+ to achieve a near maximal Po, consistent with previous observations (1). Under the conditions of our experiments, which were performed at 22-24 °C, the channel reached 50% Po at 0.8 mM calcium (Po measured at -100 mV; Fig. 7). In addition, the unitary conductance decreased with increasing [Ca2+] in the millimolar range, such that the chord conductance at -200 mV was 240 ± 9.4 picosiemens at 1 mM Ca2+ and was decreased to 89 ± 3.4 picosiemens at 25 mM Ca2+. Also consistent with previous work, we observed inward rectification in wild-type MthK (1). In symmetrical 200 mM KCl, inward current increased nearly linearly over the 0 to -200 mV range; outward current was relatively small, and saturated at 5 pA at around +100 mV for the wild-type channel.
Representative single-channel currents from wild-type and mutant MthK channels are shown in Fig. 8. As with MthK wild-type, all of the mutants that we studied displayed inward rectification (Fig. 9), which aided in determining the orientation of individual channels in the bilayer. Although some mutants also showed unitary conductances that were different from the wild type, these did not seem to correlate with the complementation results. In several of the mutants, the conductance was 12% lower than wild type (Table 1 and Fig. 9). This group included R93L, which showed robust complementation, and E92A, which resulted in almost no complementation. In addition, E96T and E96Q, which both resulted in similarly reduced complementation, showed different effects on unitary conductance, with E96Q nearly the same as wild type and E96T giving a 50% decrease.
|
We obtained recordings from three of the mutants that displayed decreased complementation. To consistently observe channel activity in our bilayer recordings, both E92Q and E96Q required liposomes with 5-fold higher protein concentrations than MthK WT liposomes (25 µg of protein/mg of lipid for E92Q and E96Q, compared with 5 µg of protein/mg of lipid for MthK WT). Despite the higher protein concentrations, channel activity was relatively low. Because of the low channel activity (quantified as NPo, the open probability times the number of channels), it was difficult to estimate the numbers of channels in each bilayer with certainty (17). Thus our estimates of channel activity (NPo = 0.14 ± 0.02 for E92Q and 0.16 ± 0.03 for E96Q) should be considered an upper limit, with the true Po for these channels likely being 3-6-fold lower. E92A, which showed the greatest reduction in complementation, also displayed a markedly reduced NPo (NPo = 0.12 ± 0.09) and required even higher protein concentrations in our liposome preparations (50 µg of protein/mg of lipid).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
Previous experiments that searched for gain-of-function mutants in KcsA identified the region near the TM2 bundle crossing as important for modulation of channel opening (10). Similarly, mutations near the putative S6 bundle crossing in Shaker can render the channel either constitutively open (e.g. P475D) or nonconducting (e.g. V478W), and modification of cysteine mutants in this region can have profound effects on gating that are consistent with direct modification of the channel's mobile gate (6, 7, 18, 19).
Although we do not presently know the structure of the closed MthK pore, sequence alignment with KcsA suggests that Glu-92 may form the narrowest point of the TM2 bundle crossing in the closed MthK channel, a point at which the Glu-92 residues from all four MthK pore-forming subunits may interact with one another directly (Fig. 1). If the Glu-92 side chains approach one another in the closed conformation, then they would be expected to repel one another electrostatically and thus destabilize the closed conformation of the putative TM2 gate. If this electrostatic mechanism were valid, then one would expect E92D and E92K also to undergo repulsion and function like the wild type. Although E92D functions like MthK wild type in the complementation assay, E92K shows a loss-of-function (Fig. 4B). This loss-of-function could be due to electrostatic repulsion of K+ by the positively charged lysine side chains at the pore entry, leading to reduced K+ conduction.
In contrast to the electrostatic interactions that may destabilize the close conformation in wild-type MthK, the four side chains at the bundle crossing in the E92A mutant are hydrophobic and may thus interact with one another more favorably in the closed state. This is consistent with the loss-of-function in E92A and the E92A/E96T double mutant (Fig. 4A). Because spontaneous channel opening is reduced in E92A at very low Ca2+ (Fig. 4), and Ca2+-dependent opening is reduced at higher Ca2+ (Fig. 8 and 9), our studies suggest that the negative charge at Glu-92 near the putative MthK bundle crossing may be critical in determining the relative stability of the MthK open state.
Our observation that the E92A mutation results in "mostly closed" channels suggests that this three-dimensional arrangement of side chains may intrinsically form a tight seal that hinders spontaneous opening. Likewise, the comparatively higher open probability with the charged Glu-92 side chain in wild-type MthK, consistent with an electrostatic destabilization of the closed conformation of the bundle crossing, is also observed with the KcsA mutants A108E and A108D (10, 16). It may thus be possible to understand gating in some K+ channels in terms of the allosteric modulation of this intrinsically stable or destabilized structure.
|
FOOTNOTES |
|---|
1 To whom correspondence should be addressed: Dept. of Physiology, MC7756, University of Texas Health Science Center, 7703 Floyd Curl Dr., San Antonio, TX 78229-3900. Tel.: 210-567-4342; Fax: 210-567-4410; E-mail: rothberg{at}uthscsa.edu.
2 The abbreviations used are: MthK, Methanobacterium thermoautotrophicum potassium channel; IPTG, isopropyl-
-D-thiogalactopyranoside; RCK, regulator of conductance of potassium; TM2, transmembrane segment 2; DM, decyl maltoside; TBS, Tris-buffered saline; WT, wild type.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  M. M.-C. Kuo, Y. Saimi, C. Kung, and S. Choe Patch Clamp and Phenotypic Analyses of a Prokaryotic Cyclic Nucleotide-gated K+ Channel Using Escherichia coli as a Host J. Biol. Chem., August 17, 2007; 282(33): 24294 - 24301. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. V. Parfenova, K. Abarca-Heidemann, B. M. Crane, and B. S. Rothberg Molecular Architecture and Divalent Cation Activation of TvoK, a Prokaryotic Potassium Channel J. Biol. Chem., August 17, 2007; 282(33): 24302 - 24309. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Kroning, M. Willenborg, N. Tholema, I. Hanelt, R. Schmid, and E. P. Bakker ATP Binding to the KTN/RCK Subunit KtrA from the K+-uptake System KtrAB of Vibrio alginolyticus: ITS ROLE IN THE FORMATION OF THE KtrAB COMPLEX AND ITS REQUIREMENT IN VIVO J. Biol. Chem., May 11, 2007; 282(19): 14018 - 14027. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. M.-C. Kuo, K. A. Baker, L. Wong, and S. Choe Dynamic oligomeric conversions of the cytoplasmic RCK domains mediate MthK potassium channel activity PNAS, February 13, 2007; 104(7): 2151 - 2156. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Li, I. Berke, L. Chen, and Y. Jiang Gating and Inward Rectifying Properties of the MthK K+ Channel with and without the Gating Ring J. Gen. Physiol., January 29, 2007; 129(2): 109 - 120. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| All ASBMB Journals | Molecular and Cellular Proteomics |
| Journal of Lipid Research | ASBMB Today |
